Oncogene studies have involved ras encoded proteins and the action of TGF-alpha on the epidermal growth factor receptor. We have made a series of chimeras between ras and the Krev- 1 gene, which encodes a ras-like protein that can suppress ras-transformed cells. The results indicated that the ras-specific and Krev-1 specific amino acids immediately surrounding residues 32-44, which are identical between the two proteins, determined whether the protein induced cellular transformation or suppressed ras transformation. Since earlier observations by us and others have implicated this region in effector function, we suggest that Krev-1 may suppress ras-induced transformation by interfering with the interaction between ras protein and its effector. The influence of GAP (GTPase activating protein) on cell transformation by ras and other oncogenes was also examined. Overexpression of GAP in NIH 3T3 cells inhibited and suppressed transformation by c-ras, c-src and an activated c-src mutant , but not by v-ras or v-mos. We conclude GAP as a negative regulator is limiting in cells and overexpression of GAP can inhibit transformation by ras-dependent oncogenes. Our results make it unlikely that GAP by itself is the ras target, but they are compatible with the possibility that GAP may be part of a complex that forms the ras target. In the studies with TGF-alpha, we have identified a system in which autocrine stimulation by TGF-alpha is more potent biologically than paracrine stimulation. Using a retroviral vector that encodes the full-length human TGF-alpha precursor protein, NIH 3T3 cells that expressed human TGF-alpha, which was processed and secreted normally, became morphologically transformed and were highly tumorigenic. By contrast to cells within the same petri dish that were not expressing TGF-alpha remained untransformed, as did cells that were treated with saturating concentrations of exogenous TGF-alpha.